How to run inference on the model

garyzhao / SemGCN

The Pytorch implementation for "Semantic Graph Convolutional Networks for 3D Human Pose Regression" (CVPR 2019).

https://arxiv.org/abs/1904.03345

Apache License 2.0

467 stars 78 forks source link

How to run inference on the model #29

Closed NielsOsnabrugge closed 4 years ago

NielsOsnabrugge commented 4 years ago

Hi @garyzhao . Thanks alot for your repo. It has been very helpfull so far. I do however have one question.

I converted my 2D prediction data to the required h36m format using your converter found here. I am however unsure how I can use this as input data to inference your pre-trained model. Is there any help you can give me to perform such an inference?

Kind Regards

NielsOsnabrugge commented 4 years ago

It seems I still had to normalize my 2D keypoints to [-1,1]. After having done this I am getting good results